Peptides have been at the forefront of combinatorial chemistry technology development due to their ease of synthesis on solid support, the reproducible and high-yielding reactions involved, and the ready availability of starting materials. Peptides are the endogenous ligands for a number of enzymes and receptors. Modifications of these peptides can be performed to develop even more potent agonists or inhibitors of these same receptors and enzymes. In addition, combinatorial peptide libraries have been used to find a number of previously unknown active sequences for a wide array of enzyme and receptor systems. However, these novel materials are still plagued by the usual limitations associated with the direct use of peptides as pharmaceuticals, although many are used in human and veterinary medicine due to their potency and selectivity. Although peptides are highly potent and selective biological agents, their use as pharmaceutical products is limited by                Poor aqueous solubility        Metabolic instability, particularly to proteases        Low oral bioavailability        Inadequate membrane permeability        Difficulty in transport to site of action in tissues and organs        Potential antigenicity        Short pharmacokinetic half-life decreases duration of pharmacological action        Side effects due to the presence of receptors for the peptide in other non-target areas of an organism        High manufacturing costs        
In order to circumvent these drawbacks while retaining the high potency of the peptide, significant work over the past three decades has been devoted to the study of mimics of these peptides, or peptidomimetics. Replacement of one or more amide bonds with functional groups that have similar structural characteristics, but different metabolic profiles has been pursued widely. Similarly, restriction of conformation of the resulting molecules utilizing either sterically demanding or structurally restricted amino acids to specifically display side chains in space. Cyclization of the linear peptide is also traditionally pursued.
However, the ability to control conformation within a single cyclic molecule often requires long experimentation in order to access the desired structure. Of greater interest would be the ability to direct and control the three-dimensional orientation so as to probe multiple conformations with the same interacting peptide side chain functionalities. In this manner, the optimal one for the biological target of interest could be rapidly determined.
Recently, WO 01/25257 has described the use of specific elements termed “tethers” to control the conformations within macrocyclic peptidomimetics. However, to date, no method has been described to combine the use of such tether elements with peptide bond surrogates.
Such molecules would have unique and superior properties over other analogues:                Ease of synthesis        Enhanced chemical stability        Improved metabolic stability        Better selectivity with lower incidence of side effects        More favorable pharmacokinetics        Better oral bioavailability        Higher aqueous solubility        
In particular, these analogues possess advantages that make them desirable as pharmaceutical agents with improved therapeutic properties:                Additional interacting functionalities        Modulated physicochemical properties        Modified conformations to those of cyclic peptides that are dictated primarily by the amide bond        
The use of backbone to backbone cyclization to control conformation of peptidic molecules has been described. (Gilon, G. Biopolymers 1991, 31, 745). However, this approach provides a constraint with the only control being provided by the length of the backbone chain employed. This does not permit access to all the conformations that might be require in order to optimally interact within a biological system. Nonetheless, this approach has yielded somatostatin analogues that can be used for therapeutic (WO 98/04583, WO 99/65508, U.S. Pat. No. 5,770,687, U.S. Pat. No. 6,051,554) or diagnostic purposes (WO 02/062819), bradykinin analogues (U.S. Pat. No. 5,874,529).
On the other hand, cyclic peptides offer a number of benefits compared with the corresponding linear analogues, including restricted conformational mobility, defined topology, enhanced stability to proteolytic enzymes and modified polarity (Molecular Diversity 2000 (pub. 2002), 5, 289-304).
Accordingly, cyclic structures can greatly improve the pharmacological and pharmacokinetic profiles of peptides. Examples demonstrate that cyclic peptides can enhance potency, selectivity, stability, bioavailability and membrane permeability. The stability to enzymatic degradation of the cyclic structure arises from the difficulty of such molecules to attain the extended conformation required to be recognized as a substrate for peptidases. Very large mixture libraries (108 members or more) of cyclic peptides have been described in WO 98/54577.
Until recently, the number of reports of the use of macrocyclic peptidomimetics in drug discovery has rather been limited. Recent examples of therapeutically interesting bioactivities that have been displayed by small peptide or peptidomimetic macrocycles include protease inhibition (HIV, cancer, inflammation)—Curr. Med. Chem. 2001, 8, 893-907; Integrin receptor antagonists (cell adhesion inhibition, inflammation, diabetes)—J. Med. Chem. 2001, 44, 2586-2592; Histone deacetylase inhibition (cancer, anti-fungal)—Tr. Endocrin. Metabol. 2001, 12, 294-300; Curr. Med. Chem. 2001, 8, 211-235; Urotensin II antagonists (cardiovascular disease)—Angew. Chem. Int. Ed. 2002, 41, 2940-2944; neurokinin-2 antagonists (asthma, irritable bowel syndrome)—J. Med. Chem. 2002, 45, 3418-3429; tyrosine receptor kinase A (TrkA) antagonists and neurotrophin-3 mimetics (Alzheimer's, stroke, diabetic neuropathy)—Mol. Pharm. 2000, 57, 385-391; J. Org. Chem. 2004; 69, 701-713; antibacterial agents—J. Med. Chem. 2002, 45, 3430-3439; and C5a complement inhibitors (inflammatory diseases)—Br. J. Pharmacol. 1999, 128, 1461-1466.
However, in most of these cases, the formation of the cyclic structure was simply one step in a lengthy optimization process. The use of large macrocyclic libraries for initial hit identification and drug discovery is largely unprecedented. This is particularly striking given the extensive efforts in combinatorial chemistry, which began focused on peptides, and the subsequent explosion in the number and type of small molecule libraries that can now be accessed.
Among the possible modifications of peptide bonds, depsipeptides are known in the art. A comparative example of a given peptide and the corresponding depsipeptide is given below. Importantly, the relative arrangement of side chains on adjacent residues is not affected as it can be with other peptide bond surrogates.

As may be noticed, one of the —NH— in the peptide is replaced by —O— in the depsipeptide.
Many depsipeptides are known to exhibit special biological activities (see Ballard, C. E.; Yu, H.; Wang, B. Curr. Med. Chem. 2002, 9, 471-498; Moore, R. E. J. Ind. Microbiol. 1996, 16, 134-143 and Shemayakin, M. M. Antimicrob. Agents Chemother 1965, 5, 962-976). For example, vancomycin, valinomycin, actinomycins, didemnins, dolstatins are natural product depsipeptides. Included in the therapeutic utility of these compounds are anticancer, antibacterial, antiviral (callipeltins, quinoxapeptins), antifungal (jaspamides), anti-inflammatory (neurokinin antagonists), anti-clotting, antiantherogenic (micropeptins), and other activities.
Another class of amino acid mimics, peptoids, have found wide utility in the design and synthesis of peptide-related therapeutic agents and biomaterials (Curr. Opin. Struct. Biol. 1999, 9, 530-535). A comparison between depsipeptides and peptoids is shown below:

In yet another approach, the urethane moiety can function as an effective peptide bond mimic. It possesses analogous defined planarity and geometry with similar rigidity to that of an amide bond. However, this moiety is not isosteric to the amide as it contains an extra atom, so that incorporation leads to larger-sized structures. This could prove quite advantageous, however, as the unique properties of peptides containing β-amino acids attests (Chem. Rev. 2001, 101, 3893-4011; Curr. Med. Chem. 2002, 9, 811-822).

The following can be cited as potential benefits of the urethane moiety as a peptide bond surrogate:                Modification of H-bonding properties due to the extra heteroatom for inter- and intramolecular interactions as well as in improved solubilities        Imposition of a degree of conformational restriction        Backbone NH and chiral R groups offer opportunities for substitution and modification for modulation of biological and physical properties        Modified polarity, more lipophilic due to the extra carbon atom, as compared with the peptide bond        Resistance to proteinases        Alteration of pharmacokinetic properties        
Urea peptide bond surrogates have also been explored in combination with other isosteres to construct molecules with novel architecture. For example, in the development of linear tripeptidomimetics as matrix metalloproteinase inhibitors for the treatment of arthritis and cancer, ureas and sulfonamides were targeted as replacements for the amide bond. The urea substitution actually contains an N-substituent where the attached group is the same as the amino acid side chain in the original peptide and hence could be considered a urea-peptoid hybrid.

These examples highlight only a representative sampling of the variety of peptide bond surrogates that have been designed and investigated (Mini-Rev. Med. Chem. 2002, 2, 463-473; Mini-Rev. Med. Chem. 2002, 2, 447-462; Curr. Med. Chem. 2002, 9, 2209-2229; Curr. Med. Chem. 2002, 9, 2243-2270; Curr. Med. Chem. 2002, 9, 963-978; Curr. Opin. Chem. Biol. 2002, 6, 872-877; Curr. Opin. Chem. Biol. 1998, 2, 441-452; Angew. Chem. Int. Ed. Engl. 1994, 33, 1699; J. Med. Chem. 1993, 36, 3039-3049; J. Org. Chem. 2000, 65, 7667-7675). Additional structures that specifically replace the peptide bond or offer an alternative type of peptide residue are shown in FIG. 3. This variety has permitted chemists to explore a number of modifications to peptide structure not accessible through natural amino acids alone. However, often this is done, not in a predictable manner, but rather determined after the construction of the molecule. Therefore, the control permitted by the aforementioned tether elements would be of utility in the context of structures containing these peptide bond surrogates.
Further, to date, peptide bond surrogates have not been widely investigated in the context of cyclic structures nor in libraries, likely due to the challenges involved in their syntheses.
Accordingly, their remains a need for macrocyclic structures incorporating a variety of peptide bond surrogates.